Cellular immunity plays a major role in controlling HIV infection in vivo. Generation of an effective vaccine against AIDS would be greatly enhanced by better understanding of the mechanisms involved in cytotoxic T-lymphocyte (CTL)-mediated immunity in HIV infection. Why CTL function fails with progression of AIDS remains largely speculative. Although some studies have suggested that failure of the CTL function in AIDS is primarily caused by functional defects in the HIV-infected CD4+ cells, direct evidence in this regard is still lacking. Functional studies for CTL response in HIV infection is generally performed using EBV-transformed autologous B cells expressing HIV antigens as the target cells. Apart from the fact that B cells are very different than CD4+ cells and are not naturally infected by HIV, there is evidence that antigen-specific response of T cells may vary depending on whether the antigens are presented endogenously or exogenously. Therefore, the CTL response against HIV antigens artificially expressed on B cells may not reflect what happens in vivo. Herpesvirus saimiri (HVS) has been used in the recent years as a powerful tool to immortalize and study human T cells. The applicants have shown that HVS can also be used to generate long-term CD4+ and CD8+ T-cell clones for functional studies from HIV-1-infected patients. Some of these CD4+ clones developed from an AIDS patient were found to be endogenously infected and spontaneously expressed HIV-1. Using these clones as target cells, the applicants found that autologous CD8+ cells failed to induce any HIV-specific CTL activity and did not inhibit HIV production. Interestingly, the same CD8+ cells were able to inhibit HIV production and demonstrated strong CTL response against other autologous HIV-negative CD4+ clones that were infected in vitro with virus produced by the endogenously infected clones. These results suggest that HIV-specific CTL activity against in vitro infected cells may not necessarily reflect the response against endogenously (in vivo) infected cells. The applicants have proposed studies to generate CD8+ and endogenously infected autologous CD4+ clones from HIV-infected patients to examine the CTL function and understand the functional defects in the in vivo infected CD4+ cells that may eventually cause failure of the CTL response.